Traditional poster

This work addresses the complexity problem inherent in local SAR estimation for parallel transmission (pTx) and shows that a relatively small subset of carefully selected “virtual observation points” is adequate for prediction and control of maximum local SAR in pTx. We tested the proposed algorithm to detect local SAR maxima by comparison with an exhaustive search over local SAR distribution in numerical simulations of adult male and female subjects for an 8 channel whole body transmit array. The proposed method of model compression for local SAR successfully captured regions of local SAR maxima, but with dramatically reduced computation cost.

Current FDA guidelines classify MR devices operating at 8 Tesla or lower as insignificant risk. Vital sign and cognitive performance data supporting the safety of performing non-proton MR imaging of the human brain at 9.4 Tesla are presented. These data add to the growing body of results that suggest ultra-high field MR imaging can be safely performed up to 9.4 Tesla.

The SAR limits issued by IEC 60601-2-33 are to some extent unbalanced comparing the whole body SAR limit with the local SAR limit. In order to examine thresholds of tissue damage a measurement setup consisting by the Tx path components of a clinical 3T MR scanner is established. Initial results of RF exposure on dead swine (6.5 W/kg, 30 minutes) are presented. Temperature increments are measured at the hot spot locations both invasively in tissue as well as superficially using infrared camera. The experimental results correspond very well with results of simulation using realistic 3D-voxel model of a swine

Influence of a receive array and failure of its detuning circuit on specific absorption rate (SAR) is investigated with simulations and experiments at 3T. A simple array of two copper loops was placed around an agar-gel phantom. A gap in the circuit was placed and removed to simulate perfect and defective detuning conditions, respectively. In simulations, addition of the ¡°decoupled¡± array resulted in only a 3% higher average SAR than with no array present, whereas with the ¡°defective¡± circuit, an increase of 41% was observed. Trends in the SAR distributions between the different cases are similar in experiment and calculation.

We have investigated the effect on power balance and SAR of simplifying rf coil 3-D EM simulation models, using co-simulation of the RF circuit and 3-D EM fields. We find that sufficiently accurate SAR and power balance estimation cannot be achieved for actual coils using simplified coil models and any renormalization approach. Simulations must converge by mesh refinement not only for B1+ values, but also for radiated and load absorbed powers, and must include simulation of the scanner gradient shield. The effect on SAR of distance to absorbing boundaries and gradient shield depends strongly on RF coil design. Thus coils and their environment need to be specified as accurately as possible, much better than current common practice. If simulation fails to give the correct power balance it is pointless to calculate SAR. In their current implementations, the Ansoft HFSS frequency domain solver provides much more reliable data, and much faster, than the CST time domain solver.

Here we compare the homogeneity in whole-brain excitation that can be achieved with two adjustable transmit channels used to drive a patch antenna and to drive a body array with elements combined as to form two orthogonal mode 1 field patterns. Optimal whole-brain homogeneity achievable with B1 shimming and the resulting SAR is examined for the two approaches separately and combined. Combining the patch and the body array results in appreciable better achievable whole-brain homogeneity with significantly lower average and maximum local SAR than possible with either the patch antenna or body coil alone.

We present a reliable and fast workflow for comprehensive joint numerical study of 7T transmit-only, receive-only RF coils as complete devices. The flexibility of the workflow allows inclusion of almost all the RF and DC components used for tune/de-tune/match/decoupling. The S-parameter data and B1+ and B1- profiles obtained for a family of loop-based 8-channel TX-only, 8-channel RX-only coils allow estimation of the required impedance of de-tuning circuits.

Estimation of local specific absorption rate (SAR) is a major challenge in high-field (>3.0 Tesla) multi-channel transmission systems. Conventional Finite-Difference Time-Domain (FDTD) method is often found inefficient to give subject-specific results. Recent studies sped up the process by applying Ampere’s law to obtain electrical field distributions via measured transverse B1 fields. The drawback is that only the longitudinal electrical fields can be calculated, which may yield large discrepancy. We present a fast numerical approach by calculating all three components of magnetic field by an integral-equation method and obtain the electric fields by Ampere’s law subsequently. Combined with fast subject modeling, this method may provide a viable approach for subject-specific SAR estimations.

Specific Absorption Rate (SAR) is a major concern in parallel transmission. Calculating the average N-gram SAR arising at each spatial location is a computationally intractable problem. We propose and demonstrate improvements to a recent algorithm, the fast region-growth algorithm, and exploit its parallelizable computation structure by implementing it on a commercial graphics card. This algorithm is up to 4-7 times faster in runtime regardless of model resolution and number of RF pulse time samples while maintaining accuracy with the fast region-growth algorithm.

We present a novel implementation of the FDTD method on a graphics card and how the possibility of simulating SAR during an MRI procedure. It is now possible to simulate the electromagnetics fields distribution in minutes or even in seconds, depending on the resolution.

The high pass birdcage body coil is commonly used in MRI for homogeneous excitation while the transverse electromagnetic volume coil is increasing used in high-field parallel transmit systems. In this work, the numerically computed normalized local SAR of four human body models, placed at three clinically relevant landmark positions, were compared for a 16-rung high-pass birdcage coil and a TEM body coil at 3T. Results show that while high local SAR may be predicted under certain conditions, any comparative generalizations of local SAR between these coils are untenable unless validated with a diverse set of human models at key landmark positions.

Ensuring RF safety is crucial for parallel transmit MRI. During parallel transmission the E-field distributions inside and outside the body exhibit a complex time evolution which cannot be covered by calorimetric SAR measurements. We introduce an optical electric field sensor (OEFS) to map non-stationary E-fields during parallel transmit MRI. We tested a customized OEFS system for certain static phase settings of a 4-channel transmit array. Our measurements confirmed a pathologic single hot-spot SAR-distribution as expected from FDTD simulations. In conclusion, the featured OEFS is an appropriate tool to identify possible safety issues of parallel transmit technologies.

Despite the considerable variations one can observe in peak local SAR among all possible driving configurations of a Tx-array, the question of how robust both SAR and flip angle distributions are with respect to small perturbations remains. A head displacement for instance not only changes the solutions of Maxwell’s equations but also the tuning of the coils. Now Cartesian feedback has been proposed as a solution to the latter problem. We therefore present the results of simulations showing the effects of small rotations of the head upon the flip angle and SAR with and without Cartesian feedback being applied.

Prediction of parallel transmission RF pulse power deposition into a patient is one of the principal challenges at ultra-high field magnetic strength. We introduced real time calibration system to predict the net power consequences of the parallel transmission RF pulse design before employing designed RF pulses into scanner. High accuracy was achieved in the power deposition estimation of designed RF pulses with respect to actual net power measurements.

An anatomically accurate porcine head model was developed. The model was developed to help validate the temperature predictions of bioheat equations against direct in vivo fluoroptic measurements and predict non-uniform brain RF heating in swine and humans wearing/not-wearing implantable, conductive medical devices for a variety of field strengths, coil configurations, and head loading positions. The head model was developed by obtaining high resolution images of a porcine head using a Siemens 3T Trio (1.02 mm X 1.02 mm X 1.00 mm, Sequence type: T1MPRAGE) and manually segmenting the brain, cerebral spinal fluid, bone, cartilage, muscle, and air-cavity using MIMICS.

We present a Bloch-based MRI simulator that considers desired sample, field distributions (B0, B1, E1, Gx, Gy, Gz) and pulse sequence for determining effects of field distributions on images (including noise correlation) and SAR. The software is provided with a few basic libraries for simulation of head and body in 8-element transceiver arrays at 3T and 7T and a GUI for designing and executing some simple pulse sequences, but structure of input files is intentionally simple so users can generate their own sample, field distributions, and sequence files. Many fundamental capabilities are demonstrated.

Vertebrate animal MRI is an important part of medical research, and veterinary MRI imaging of companion animals is increasing. Human subjects are generally provided with hearing protection against the loud, potentially damaging acoustic noise produced by MRI scanners; this is generally not done for animal MRI subjects. Hearing damage can interfere with research functions for research animals or quality of life for companion animals. We compare typical MRI noise levels to animal hearing thresholds and conclude that MRI exposes many animals to levels of noise and duration that would exceed NIOSH limits for human exposure.

Subjective acceptance of ultra-high-field MRI has not been evaluated in larger study groups yet. For this purpose, 573 volunteers underwent a 7T examination and were afterwards asked about sensations and side effects. Analysis revealed an overall high subjective acceptance of 7T examinations with mainly non-specific factors such as unpleasant room temperature, little contact to the staff or noise being criticized. Compared to 1.5T volunteers described considerably more often nausea or a metallic taste on 7T; however, the average degree of these effects was very low. Volunteers lying ”head first” expressed more complaints than those lying “feet first”.

In this study, a new gradient pulse, Dynamic Slew Rate Pulse (DSRP), is designed to reduce peripheral nerve stimulation (PNS). Compared with traditional trapezoid pulse, DSRP¡¯s pulse width is much smaller when dB/dt limitation is dominant.

Due to the insufficient experimental data on static magnetic field exposure and new ICNIRP limitations an exposure measurement instrument was developed. With the new probe actual B₀ and time varying magnetic fields due to movement in the stray field can be detected simultaneously. Therefore, personal exposure of 10 healthcare workers and MR physicists during normal operating procedures could be determined. Peak exposure limits are exceeded during all procedures.

We demonstrate that active noise control (ANC) is feasible to treat MRI noises at frequencies greater than 2 KHz. Several MRI scan noise and gradient signals were recorded and replayed in a sound quality chamber to simulate the MRI environment. A dummy “wore” headphones containing piezoceramic speakers with condenser microphones installed inside and outside the earpiece to measure the environmental sound in the immediate patient vicinity. During the simulation study, the sound pressure level (SPL) was measured, both with and without the ANC. Results presented show the ANC system attained significant SPL reduction at all frequencies up to 5 kHz.

Recently, new approaches for application of MRI in various branches of dentistry have been proposed. Dental materials present in the subject’s mouth pose a major concern for dental applications of MRI. Partly contradictory results have been reported regarding the severity of image artifacts caused by different dental materials, usually without consideration of dental applications of MRI. This paper provides classification of standard dental materials from the standpoint of dental MRI, and can serve as a guideline in future dental MRI research.

While imaging subjects in an MRI scanner, it may sometimes be necessary to de-fibrillate in order to restore sinus rhythm or resuscitate. Normal de-fibrillator equipment is incompatible with the scanner's magnetic field. Presented is a procedure for de-fibrillating inside the MRI environment. With this procedure, the defibrillator unit was located outside the MRI environment, with select non-ferrous equipment inside and on the subject. This procedure was employed on four separate occasions with an animal subject on the scanner table, and was found to be successful. No problems or safety concerns were observed.

While MR is a diagnostic imaging tool which saves lives, magnetic forces of fringe magnetic field components of MR systems on ferromagnetic components can impose a severe occupational health and safety hazard. With the advent of ultrahigh field MR Systems – including passively shielded magnet versions – this risk - commonly known as missile effect /1/ - is pronounced. Numerous accidents have been reported ranging from mechanical damage to patient death. These casualties are probably most widely known through television documentaries and printed media but still present the tip of the iceberg of safety violations (/2/ /3/ /4/). Various policies have been implemented around the world to safeguard healthcare workers, volunteers and patients with the ultimate goal of avoiding unforeseen disasters and injuries.